TEKDENCT OF SPECIES TO FOBM VAllTETIES. 61 



the preservation of the individual animals (without any thought 

 of their offspring) most useful to each race of man in his par- 

 ticular circumstances. The "roguing," as nurserymen call the 

 destroying of varieties which depart from their type, is a kind of 

 selection. I am convinced that intentional and occasional selec- 

 tion has been the main agent in the production of our domestic 

 races ; but however tliis may be, its great power of modification 

 has been indisputably shown in later times. Selection acts only 

 by the accumulation of slight or greater variations, caused by ex- 

 ternal conditions, or by the mere fact that in generation the child 

 is not absolutely similar to its parent. Man, by this power of 

 accumulating variations, adapts living beings to his wants — may 

 be said to make the wool of one sheep good for carpets, of another 

 for cloth, &c. 



2. Now suppose there were a being who did not judge by mere 

 external appearances, but who could study the whole internal or- 

 ganization, who was never capricious, and should go on selecting 

 for one object during millions of generations ; wlio will say what he 

 might not effect ? In nature we have some slight variation occa- 

 sionally in all parts ; and I think it can be shown that changed 

 conditions of existence is the main cause of the child not exactly 

 resembling its parents ; and in nature geology shows us what 

 changes have taken place, and are taking place. We have almost 

 unlimited time ; no one but a practical geologist can fully appre- 

 ciate this. Think of the Glacial period, during the whole of which 

 the same species at least of shells have existed ; there must have 

 been during this period millions on millions of generations. 



3. I think it can be shown that there is such an unerring 

 power at work in Natural Selection (the title of my book), which 

 selects exclusively for the good of each organic being. The elder 

 De Candolle,'W. Herbert, and Lyell have written excellently on the 

 struggle for life ; but even they have not written strongly enough. 

 Eeflect that every being (even the elephant) breeds at such a rate, 

 that in a few years, or at most a few centuries, the surface of the 

 earth would not hold the progeny of one pair. I have found it 

 hard constantly to bear in mind that the increase of every single 

 species is checked during some part of its life, or during some 

 shortly recurrent generation. Only a few of those annually born 

 can live to propagate their kind. What a trifling difference must 

 often determine which shall survive, and which perish ! 



4. Now take the case of a country undergoing some change. 

 This will tend to cause some of its inhabitants to vary slightly — 



4* 



